The aim of this research is to understand how stable states of gene expression are maintained during development. Molecular, genetic, and biochemical methods will be used to study proteins that control transcription of homeotic genes in Drosophila melanogaster. The homeotic genes determine the identities of segments in the fly. Expression of homeotic genes, in restricted patterns along the anterior-posterior axis, is needed throughout embryonic, larval and pupal development. Segmentation gene products control the initial Polycomb group (PcG) products are transcriptional repressors that then maintain restricted homeotic patterns throughout the rest of development. Current models suggest that PcG products recognize the initial off states of homeotic genes established in early embryos and they maintain repression by assembling stable protein- DNA complexes in the local chromatin. This research will investigate the individual roles of PcG repressor proteins. The work will focus on two of the PcG products, extra sex combs (esc) protein and Sex comb on midleg (Scm) protein, which are likely to play different roles in repression. esc may play a short-term role in assembling or guiding other objectives of this work are to define the functional domains in esc and Scm and to determine if and how they interact with other PcG proteins. Site-specific mutagenesis, coupled with germline transformation and transient rescue assays, will test the role of WD40 repeats in esc protein, which are likely used for protein-protein contact. DNA sequence analysis of mutant alleles and site-directed mutagenesis will be used to test the functional domains in Scm, including several zinc fingers. Isolation of esc and Scm homologs in other insect species will be used to identify functional domains. These studies will be extended to test for mammalian homologs of these PcG genes. Gel shift assays will test if Scm protein binds to DNA. In vitro binding assays, yeast wi-hybrid tests, and co-immune precipitations will check for physical interactions between esc, Scm and other PcG proteins. These interactions will also be tested in vivo using PcG proteins fused to a DNA-binding domain. esc and Scm fusion proteins will be tethered to specific sites in the genome and chromosome immunostaining will test for co-association of other PcG proteins. This work will address how cell fate decision are maintained during fly development. In addition, the mammalian Hox genes control anterior- posterior patterning in ways that resemble the fly homeotic genes. There are several mouse genes that are related to fly PcG genes, including at least one that is a functional homolog. Thus, an understanding of PcG repression in files should provide clues about similar controls in higher organisms. Since the mouse PcG homolog is an oncogene, this work may also provide insight into the basis of uncontrolled cell proliferation.